Qianfan OCR needs about 4.4 GB of memory at Q4_K_M (a 2.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 11 quants, from 1.8 GB (Q2_K) to 8.8 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.
4B-parameter end‑to‑end OCR model
Qianfan OCR is a 4.42B‑parameter vision‑language model that performs direct image‑to‑Markdown conversion. It unifies document parsing, layout analysis, and understanding in a single architecture. The model supports reasoning, tools, and vision tasks.
Good at
- End‑to‑end OCR
- Document parsing
- Layout analysis
Limits
- No vision support in some quantizations
- Large VRAM needed for high context
- License is apache‑2.0
Best for
- Document digitization
- Table extraction
- Key information extraction
Who it’s for. Developers building document intelligence pipelines who need a single model for OCR, layout, and understanding. The model is suitable for applications requiring multilingual OCR and structured output.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 1.8 GB | 3.5 GB | Fits | Get file → | |
| Q3_K_S | 2.1 GB | 3.8 GB | Fits | Get file → | |
| Q3_K_M | 2.2 GB | 4.0 GB | Fits | Get file → | |
| Q4_0 | 2.6 GB | 4.3 GB | Fits | Get file → | |
| Q4_K_S | 2.6 GB | 4.3 GB | Fits | Get file → | |
| Q4_K_M | 2.7 GB | 4.4 GB | Fits | Get file → | |
| Q5_K_S | 3.1 GB | 4.8 GB | Fits | Get file → | |
| Q5_K_M | 3.2 GB | 4.9 GB | Fits | Get file → | |
| Q6_K | 3.6 GB | 5.3 GB | Fits | Get file → | |
| Q8_0 | 4.7 GB | 6.4 GB | Fits | Get file → | |
| BF16 | 8.8 GB | 10.6 GB | Too big | Get file → |
For 8 GB VRAM use Q4_K_M (file size 2.72 GB, estimated 4.4 GB memory). For 12 GB VRAM use Q5_K_M (file size 3.16 GB, estimated 4.9 GB memory). For 16 GB VRAM use Q6_K (file size 3.63 GB, estimated 5.3 GB memory).
What GPU do you need for Qianfan OCR?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q3_K_M (2.2 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K (3.6 GB file, about 5.3 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (4.7 GB file, about 6.4 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (8.8 GB file, about 10.6 GB in use).
Which Qianfan OCR quant should you choose?
Q4_K_M is the usual starting point for Qianfan OCR: it balances file size and quality well. Here it is a 2.7 GB file.
The Q8_0 file is 4.7 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 1.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qianfan OCR
How much VRAM does Qianfan OCR need?
About 4.4 GB at Q4_K_M with an 8K context window. That is the 2.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Qianfan OCR on an 8 GB GPU?
Yes, with Q8_0: the 4.7 GB file needs about 6.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Qianfan OCR GGUF?
It depends on the quant: Q2_K is 1.8 GB, Q3_K_S is 2.1 GB, Q3_K_M is 2.2 GB, Q4_0 is 2.6 GB, Q4_K_S is 2.6 GB, Q4_K_M is 2.7 GB and 5 more.
What context length does Qianfan OCR support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.
Where can I download Qianfan OCR GGUF?
The files are on Hugging Face in the Reza2kn/Qianfan-OCR-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Qianfan OCR use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Qianfan OCR on 8 GB VRAM?
Yes, the Q4_K_M quantization requires approximately 4.4 GB VRAM at 8 k context, which fits within 8 GB.
Does the model support vision tasks?
Yes, the model includes vision capabilities as part of its multimodal architecture.
What license applies to the GGUF files?
The GGUF distribution is licensed under Apache‑2.0, matching the upstream model.